Repent and Believe in the Data!
There’s a new trend among famous economists, from Krugman to Cowen. It was started by Adam Ozimek with a simple question: "Has a single economics study changed your mind on an important issue?" Economics as a field of study has been branded “too ideological” with warring camps defending dogmatic fortresses as opposed to unbiased, independent scholars carefully testing theories and reporting their results and methods for replication by others. As a result, many economists are answering Ozimek and pointing out the times they changed their views in light of new findings.
The Economist says they are defending the profession, but the trend really shows just how unscientific and ideological mainstream economics has become.
Economics and Science
According to The Economist, the trend is just one more step toward the goal of unbiased, non-ideological economic science:
But even if economics is not uniquely ideological, its biases are often more salient than those within chemistry. Economists advise politicians on all manner of important decisions. A reputation for impartiality could improve both perceptions of the field and the quality of economic policy.
Of course many of these mind-changing moments involve dropping even the simplest and fundamental results in economics for platforms for increased state intervention. Krugman, for example, cites his turn on minimum wage laws after he read the famous empirical paper by Card and Krueger and others. He doesn’t explain why the standard theory is wrong, just that it must be wrong because the data say so.
This would all be fine if economics were a natural science like physics or chemistry. The subject matter for these sciences cannot choose their behavior—we explain the behavior of rocks and molecules using constants and fundamental laws of the universe, which are discovered by observing regularities in the way the rocks and molecules and whatnot behave.
We wouldn’t say that a rock chooses to fall if we drop it from some height, or combustion tries to make carbon dioxide. Indeed, these claims would be labeled very unscientific because they don’t comport with reality. Laboratory experiments and recording empirical regularities are therefore well-suited for these sciences because of their subject matter.
The Logic of Action
Economics is also a science, but the subject matter for economics is categorically different. Humans can choose. This very fact actually serves as the basis for the entire structure of economic theory. Recognizing this point is critical for selecting and preserving an appropriate method for economics.
When we choose, we attempt to satisfy some end using some means. We must select something over something else, and so our ends can be conceived as existing in a list, or a ranking. As such, every action has some opportunity cost—some other end could have been satisfied using the same means.
Proceeding further, we can derive the law of demand, the law of supply, the existence of market prices, the effects of the division of labor, the structure of production, and the effects of all sorts of events and policies.
All of this can be done with logic and thought experiments that isolate cause and effect, and with the help of only a few very safe assumptions. The point is, this logical-deductive method is most appropriate for economics because of the subject matter: acting, choosing individuals.
This uniquely Austrian position is sometimes attacked as contrived or disconnected from the real world, when it actually connects with the real world and real humans on a much deeper level than treating humans like rocks or choice like a chemical reaction.
The Place for Data in Economics
Laboratory experiments and real-world observation cannot somehow refute economic theory because economics deals in counterfactuals, which are never realized and cannot be observed in the real world.
This is not to say that good economists should ignore data. In fact, a good economist would revisit theory in the face of curious data that doesn’t seem to go along with theory. Also, economic history relies on the skillful use of data and theory to help explain events of the past.
The difference is that new theory should not be based on empirical findings. Consistently “weird” empirical findings might encourage good economists to take another look at their theory, however. In addition, good economists have a healthy skepticism of data, especially when it seems to go against the grain of long-established theory.
Economic Ignorance, not Enlightenment
So, this new trend among prominent economists, far from defending economics as a science and a profession, actually reveals the unscientific and ideological proclivities of mainstream economics.
Readily dropping standard results in economics because the newest econometric trick revealed something strange with the newest data from the Bureau of Labor Statistics shows economic ignorance, not enlightenment. It shows that many economists don’t understand the fundamentals of economics and the logic of action.
When data conflicts with theory, the scholarly thing to do would be to find the source of the disconnect. This could be an error in the way the data were measured or an error in the application of the data to a specific economic theory. Of course, the theory itself could be wrong, too. Either way, some explanation is in order. An ideologue would discard one or the other without any reason except to maintain their own ideology or to switch ideological camps.